Objectives Malignant external otitis (MEO) is a potentially fatal infection of the external auditory canal, temporal bone, and skull base. Despite treatment with modern antibiotics, MEO can lead to skull base osteomyelitis. Until now, there have been few studies on the prognostic factors of MEO.Methods We performed a retrospective study to identify prognostic factors of MEO, and a meta-analysis of other articles investigating MEO. On the basis of disease progression the 28 patients in our study were divided into ‘controlled’ and ‘uncontrolled’ groups, consisting of 12 and 16 patients, respectively. We identified three categories of prognostic factors: those related to patient, disease, and treatment. We compared these prognostic factors between the controlled and uncontrolled groups.ResultsIn our study, the duration of diabetes mellitus (DM), presence of inflammatory markers (C-reactive protein and erythrocyte sedimentation rate), and computed tomography or magnetic resonance imaging findings influenced the prognosis of MEO. In contrast, prognosis was unrelated to age, gender, mean glucose level, hemoglobin A1c level, pathogen, comorbidity, or cranial nerve involvement. No factor related to treatment modality was correlated with prognosis, such as surgery, steroid therapy, or interval to the first appropriate treatment. Cranial nerve involvement has been proven to be associated with disease progression, but the relationship between cranial nerve involvement and the prognosis of MEO remains controversial. As a part of this study, we conducted a meta-analysis of cranial nerve involvement as a prognostic factor of MEO. We found that cranial nerve involvement has a statistically significant influence on the prognosis of MEO.Conclusion We found that glycemic control in diabetes mellitus, cranial nerve involvement, and the extent of disease determined from various imaging modalities influence the prognosis of MEO. We suggest that significant prognostic factors should be monitored to determine the prognosis of patients with MEO.

A definitive study on the prevalence of adult unilateral hearing loss and hearing aid rehabilitation is lacking in Korea. The purpose of our study was to investigate the prevalence of adult unilateral hearing loss and the factors associated with hearing aid use in patients with unilateral hearing loss in South Korea. We obtained data from 2009 to 2012 from the Korea National Health and Nutrition Examination Surveys (KNHANES), a cross-sectional, nationwide and population-based survey in the Republic of Korea. We analyzed the prevalence and associated factors of unilateral hearing loss and hearing aid adoption by univariable and multivariable analysis. Unilateral hearing loss was defined as pure tone average � 41 dB in the worse hearing ear, and < 41 dB in the other ear assessed at 0.5, 1.0, 2.0, and 3.0 kHz. From 2009 to 2012, 33,252 individuals participated in the KNHANES. Among them, the number of patients with unilateral hearing loss was 1632 (5.55%) and the prevalence of hearing aid adoption in unilateral hearing loss was 1.56%. We also compared the factors between hearing aid users and non-users. Occupational status (OR 3.759, 95% CI 1.443-9.804), the hearing threshold in the better ear (OR 1.088, 95% CI 1.029-1.151), and hearing threshold in the worse ear (OR 1.031, 1.005-1.058) were found to affect the adoption of hearing aids. The prevalence of noise exposure at work in hearing aid users was significantly lower than the prevalence of noise exposure at work in those with no hearing aid. The prevalence of hearing aid use in patients with unilateral hearing loss in Korea is very low compared to other countries. Public health education is needed to increase public awareness of unilateral hearing loss, hearing aid adoption and its continued usage. Auditory rehabilitation should be actively recommended to patients with unilateral hearing loss.

Background Deep learning (DL)–based artificial intelligence may have different diagnostic characteristics than human experts in medical diagnosis. As a data-driven knowledge system, heterogeneous population incidence in the clinical world is considered to cause more bias to DL than clinicians. Conversely, by experiencing limited numbers of cases, human experts may exhibit large interindividual variability. Thus, understanding how the 2 groups classify given data differently is an essential step for the cooperative usage of DL in clinical application. Objective This study aimed to evaluate and compare the differential effects of clinical experience in otoendoscopic image diagnosis in both computers and physicians exemplified by the class imbalance problem and guide clinicians when utilizing decision support systems. Methods We used digital otoendoscopic images of patients who visited the outpatient clinic in the Department of Otorhinolaryngology at Severance Hospital, Seoul, South Korea, from January 2013 to June 2019, for a total of 22,707 otoendoscopic images. We excluded similar images, and 7500 otoendoscopic images were selected for labeling. We built a DL-based image classification model to classify the given image into 6 disease categories. Two test sets of 300 images were populated: balanced and imbalanced test sets. We included 14 clinicians (otolaryngologists and nonotolaryngology specialists including general practitioners) and 13 DL-based models. We used accuracy (overall and per-class) and kappa statistics to compare the results of individual physicians and the ML models. Results Our ML models had consistently high accuracies (balanced test set: mean 77.14%, SD 1.83%; imbalanced test set: mean 82.03%, SD 3.06%), equivalent to those of otolaryngologists (balanced: mean 71.17%, SD 3.37%; imbalanced: mean 72.84%, SD 6.41%) and far better than those of nonotolaryngologists (balanced: mean 45.63%, SD 7.89%; imbalanced: mean 44.08%, SD 15.83%). However, ML models suffered from class imbalance problems (balanced test set: mean 77.14%, SD 1.83%; imbalanced test set: mean 82.03%, SD 3.06%). This was mitigated by data augmentation, particularly for low incidence classes, but rare disease classes still had low per-class accuracies. Human physicians, despite being less affected by prevalence, showed high interphysician variability (ML models: kappa=0.83, SD 0.02; otolaryngologists: kappa=0.60, SD 0.07). Conclusions Even though ML models deliver excellent performance in classifying ear disease, physicians and ML models have their own strengths. ML models have consistent and high accuracy while considering only the given image and show bias toward prevalence, whereas human physicians have varying performance but do not show bias toward prevalence and may also consider extra information that is not images. To deliver the best patient care in the shortage of otolaryngologists, our ML model can serve a cooperative role for clinicians with diverse expertise, as long as it is kept in mind that models consider only images and could be biased toward prevalent diseases even after data augmentation.

Objectives Inlay butterfly cartilage tympanoplasty (IBCT) is a simple grafting technique. Endoscopy facilitates visualization by eliminating blind spots. We analyzed the outcomes of IBCT using both endoscopic and microscopic approaches, and assessed how trainees perceived the educational opportunities afforded. Materials and methods Sixty patients who underwent IBCT were allocated to Group I (n = 30; microscopic IBCT) and Group II (n = 30; endoscopic IBCT) by the dates of their visits. Anatomical success was defined as an intact, repaired tympanic membrane; functional success was defined as a significant decrease in the air–bone gap. Postoperative discomfort was analyzed using a visual analog scale (VAS). Thirteen trainees completed structured questionnaires exploring anatomical identification and the surgical steps. Results The surgical success rates were 96.7% in Group I and 100% in Group II. We found no between-group differences in the mean decrease in the air–bone gap or the extent of postoperative discomfort. Significant postoperative hearing improvements were evident in both groups. The mean operative time was shorter when the microscopic approach was chosen (17.7±4.53 vs. 26.13±9.94 min). The two approaches significantly differed in terms of the identification of external and middle ear anatomical features by the trainees, and their understanding of the surgical steps. Conclusion Both endoscopic and microscopic IBCT were associated with good success rates. The endoscopic approach facilitates visualization, and a better understanding of the middle ear anatomy and the required surgical steps and thus is of greater educational utility.

Background and Objectives Inlay butterfly cartilage tympanoplasty makes the graft easy, and reduces operating time. The present study aimed to investigate the outcomes of microscopic versus endoscopic inlay butterfly cartilage tympanoplasty. Subjects and Methods In this retrospective study, the outcomes of 63 patients who underwent inlay butterfly cartilage tympanoplasty with small to medium chronic tympanic membrane perforation were evaluated. Twenty-four patients underwent conventional microscopic tympanoplasty and 39 underwent endoscopic tympanoplasty. The outcomes were analyzed in terms of the hearing gain and graft success rate. Results The surgical success rate was 95.8% in the patients who underwent conventional microscopic tympanoplasty and 92.3% in those who underwent endoscopic tympanoplasty. In both groups of patients, the postoperative air-bone gap (ABG) was significantly lower than the preoperative ABG. There were no significant differences between the preoperative and postoperative ABG values in either group. Conclusions Endoscopic inlay tympanoplasty using the butterfly cartilage technique appears to be an effective alternative to microscopic tympanoplasty and results in excellent hearing.